Abstract/Project Summary Cellular RNAs differ in sequence, structure, and function from their precursor RNAs. The enormous regulatory power of pre-mRNA processing is influenced by transcription, because the RNA processing machinery acts co-transcriptionally and can contact RNA polymerase II (Pol II) and/or chromatin directly. In turn, each step in pre-mRNA processing ? 5' end capping, splicing, and 3' end cleavage ? is associated with changes in Pol II behavior, such as pausing. This grant focuses on the coordination between transcription, splicing, and 3' end cleavage in Saccharomyces cerevisiae and Schizosaccharomyces pombe. Our findings during the past 3 years have redefined how we think about the cross-talk between these three aspects of mRNA maturation. We developed two complementary single molecule RNA-seq methods that directly measure the progression of the splicing reaction as a function of the position of elongating Pol II on a global scale, resulting in two major discoveries: (1) We have shown that the spliceosome operates on a much faster time scale than previously appreciated and is close to Pol II when it acts. The short lag between splicing and transcription raises intriguing questions about how splicing is coordinated with transcription, mRNP maturation, and 3' end cleavage at gene termini. In Aim 1, we test models regarding the timing of spliceosome assembly in response to RNA sequence, RNA structure, and the action of trans-acting factors. In Aim 2, we investigate a novel hypothesis ? that removal of the post- catalytic spliceosome from nascent mRNA, which we call spliceosome eviction, is necessary to promote mRNP maturation at sites of Pol II pausing ? and experimentally evaluate this idea against other models. (2) One of our methods, long read sequencing, determines the full sequence of nascent RNA from 5' to 3' end (Pol II position), enabling us to track the splicing of abundant multi-intron transcripts in S. pombe and determine the order of co-transcriptional intron removal in ?real time?. Remarkably, most nascent transcripts were spliced in an ?all or none? fashion, such that more than half were rapidly and fully spliced. In contrast, 18% of nascent transcripts were totally unspliced, failed to undergo 3' end cleavage, and were degraded by the nuclear exosome. These ?dead-end? transcripts display transcriptional readthrough, which has recently been implicated in cellular responses to stress, cancer, and viral infection. In Aim 3 we propose to identify the molecular mechanisms that define transcript fate with regard to splicing, 3' end cleavage, and degradation. The impact of our new aims will be to define the repertoire of intron features that determine the kinetics of spliceosome assembly in vivo; identify coordinated transitions in transcription and mRNP maturation; and discover how transcription, splicing and 3' end cleavage are linked for success or failure in the context of normal growth and cellular stress.